Video Display Hardware

Your monitor provides the link between you and your computer. Although you can possibly get rid of your printer, disk drives, and expansion cards, you cannot sacrifice the monitor. Without it, you would be operating blind; you could not see the results of your calculations or the mistyped words on-screen.

The first microcomputers were small boxes that lacked displays. Instead, users observed the information contained in system registers via banks of flashing LEDs and waited for the final output to be printed. All interaction with the system was normally done through a typewriter terminal. When the CRT (Cathode Ray Tube) terminal or monitor was finally added as an interface, the computer became more attractive to a wider audience. This visual trend in user interface technology continued later with the adoption of graphical user interfaces such as Windows over text-based systems like DOS.

The video subsystem of a PC consists of two main components:

• Monitor (or video display)
  
  
• Video adapter (also called the video card or graphics card)

This chapter explores the range of different PC-compatible video adapters and the displays that work with them.

Monitors

The monitor is, of course, the display located on top of, near, or inside your computer. Like any computer device, a monitor requires a source of input. The signals that run to your monitor come from video circuitry inside or plugged into your computer. Newer computers--such as those that use the Low Profile (LPX or NLX) motherboard form factor--usually contain this circuitry on the motherboard. Other systems incorporate the video on a separate circuit board that is plugged into an expansion or bus slot. The expansion cards that produce video signals are called video cards, video adapters, or graphics cards. Whether the video circuit is built into the motherboard or on a separate card, the circuitry operates the same way and uses generally the same components.

Display Technologies

A monitor may use one of several display technologies. By far the most used is cathode ray tube (CRT) technology, the same technology used in television sets. CRTs consist of a vacuum tube enclosed in glass. One end of the tube contains an electron gun; the other end contains a screen with a phosphorous coating.

When heated, the electron gun emits a stream of high-speed electrons that are attracted to the other end of the tube. Along the way, a focus control and deflection coil steer the beam to a specific point on the phosphorous screen. When struck by the beam, the phosphor glows. This light is what you see when you watch TV or your computer screen.

The phosphor chemical has a quality called persistence, which indicates how long this glow will remain on-screen. You should have a good match between persistence and scanning frequency so that the image has less flicker (if the persistence is too low) and no ghosts (if the persistence is too high).

The electron beam moves very quickly, sweeping the screen from left to right in lines from top to bottom, in a pattern called a raster. The horizontal scan rate refers to the speed at which the electron beam moves across the screen.

During its sweep, the beam strikes the phosphor wherever an image should appear on-screen. The beam also varies by intensity in order to produce different levels of brightness. Because the glow fades almost immediately, the electron beam must continue to sweep the screen to maintain an image--a practice called redrawing or refreshing the screen.

Most older displays have a maximum refresh rate (also called a vertical scan frequency) of about 70 hertz (Hz), meaning that the screen is refreshed 70 times a second. Newer displays usually have higher refresh rates. Low refresh rates cause the screen to flicker, contributing to eye strain. The higher the refresh rate, the better for your eyes.

It is important that the scan rates expected by your monitor match those produced by your video card. If you have mismatched rates, you cannot see an image and may actually damage your monitor.

Some monitors have a fixed refresh rate. Other monitors may support a range of frequencies; this support provides built-in compatibility with many video standards (described in the "Video Cards" section later in this chapter). A monitor that supports many video standards is called a multiple-frequency monitor. Most newer monitors are multiple-frequency monitors, which means that they support operation with a variety of popular video signal standards. Different vendors call their multiple-frequency monitors by different names, including multisync, multifrequency, multiscan, autosynchronous, and autotracking.

Phosphor-based screens come in two styles--curved and flat. The typical display screen is curved, meaning that it bulges outward from the middle of the screen. This design is consistent with the vast majority of CRT designs (the same as the tube in your television set).

The traditional screen is curved both vertically and horizontally. Some models use the Trinitron design, which is curved only horizontally and is flat vertically. Many people prefer this flatter screen because it results in less glare and a higher-quality, more accurate image. The disadvantage is that the technology required to produce flat-screen displays is more expensive, resulting in higher prices for the monitors.

Alternative display designs are available. Borrowing technology from laptop manufacturers, some companies provide LCD (liquid-crystal display) displays. LCDs have low-glare flat screens and low power requirements (5 watts versus nearly 100 watts for an ordinary monitor). The color quality of an active-matrix LCD panel actually exceeds that of most CRT displays. At this point, however, LCD screens usually are more limited in resolution than typical CRTs and are much more expensive. There are three basic LCD choices: passive-matrix monochrome, passive-matrix color, and active-matrix color. The passive-matrix designs are also available in single- and dual-scan versions.

In an LCD, a polarizing filter creates two separate light waves. In a color LCD, there is an additional filter that has three cells per each pixel--one each for displaying red, green, and blue.

The light wave passes through a liquid-crystal cell, with each color segment having its own cell. The liquid crystals are rod-shaped molecules that flow like a liquid. They enable light to pass straight through, but an electrical charge alters their orientation, as well as the orientation of light passing through them. Although monochrome LCDs do not have color filters, they can have multiple cells per pixel for controlling shades of gray.

In a passive-matrix LCD, each cell is controlled by electrical charges transmitted by transistors according to row and column positions on the screen's edge. As the cell reacts to the pulsing charge, it twists the light wave, with stronger charges twisting the light wave more. Supertwist refers to the orientation of the liquid crystals, comparing on mode to off mode--the greater the twist, the higher the contrast.

Charges in passive-matrix LCDs are pulsed, so the displays lack the brilliance of active-matrix, which provides a constant charge to each cell. To increase the brilliance, some vendors have turned to a new technique called double-scan LCD, which splits passive-matrix screens into a top half and bottom half, cutting the time between each pulse. Besides increasing the brightness, dual-scan designs also increase the response time or speed of the display, making this type more usable for video or other applications where the displayed information changes rapidly.

In an active-matrix LCD, each cell has its own transistor to charge it and twist the light wave. This provides a brighter image than passive-matrix displays because the cell can maintain a constant, rather than momentary, charge. However, active-matrix technology uses more energy than passive-matrix. With a dedicated transistor for every cell, active-matrix displays are more difficult and expensive to produce.

In both active- and passive-matrix LCDs, the second polarizing filter controls how much light passes through each cell. Cells twist the wavelength of light to closely match the filter's allowable wavelength. The more light that passes through the filter at each cell, the brighter the pixel.

Monochrome LCDs achieve gray scales (up to 64) by varying the brightness of a cell or dithering cells in an on-and-off pattern. Color LCDs, on the other hand, dither the three-color cells and control their brilliance to achieve different colors on the screen.

The big problem with active-matrix LCDs is that the manufacturing yields are low, forcing higher prices. This means that many of the panels produced have more than a certain maximum number of failed transistors. The resulting low yields limit the production capacity and incur higher prices.

In the past, several hot CRTs were needed to light an LCD screen, but portable computer manufacturers later used a single tube the size of a cigarette. Light emitted from the tube gets spread evenly across the entire display using fiber-optic technology.

Thanks to supertwist and triple-supertwist LCDs, newer screens enable you to see the screen clearly from more angles with better contrast and lighting. To improve readability, especially in dim light, some laptops include backlighting or edgelighting (also called sidelighting). Backlit screens provide light from a panel behind the LCD. Edgelit screens get their light from the small fluorescent tubes mounted along the sides of the screen. Some older laptops excluded such lighting systems to lengthen battery life. Most laptops enable you to run the backlight at a reduced power setting that dims the display but allows for longer battery life.

The best color displays are active-matrix or Thin-Film Transistor (TFT) panels, in which each pixel is controlled by three transistors (for red, green, and blue). Active-matrix-screen refreshes and redraws are immediate and accurate, with much less ghosting and blurring than in passive-matrix LCDs (which control pixels via rows and columns of transistors along the edges of the screen). Active-matrix displays are also much brighter and can easily be read at an angle.

An alternative to LCD screens is gas-plasma technology, typically known for its black and orange screens in some of the older Toshiba notebook computers. Some companies use gas-plasma technology for desktop screens and color high-definition television (HDTV) flat-panel screens.

Monochrome versus Color

During the early years of the IBM PC and compatibles, owners had only two video choices--color using a CGA display adapter and monochrome using an MDA display adapter. Since then, many adapter and display options have hit the market.

Monochrome monitors produce images of one color. The most popular is amber, followed by white and green. The color of the monitor is determined by the color of the phosphors on the CRT screen. Some monochrome monitors with white phosphors can support many shades of gray.

Color monitors use more sophisticated technology than monochrome monitors. Whereas a monochrome picture tube contains one electron gun, a color tube contains three guns arranged in a triangular shape referred to as a delta configuration. Instead of amber, white, or green phosphors, the monitor screen contains phosphor triads, which consist of one red phosphor, one green phosphor, and one blue phosphor arranged in the same pattern as the electron guns. These three primary colors can be mixed to produce all other colors.

The Right Size

Monitors come in different sizes, ranging from 9-inch to 42-inch diagonal measure. The larger the monitor, the higher the price tag. The most common monitor sizes are 14, 15, 17, 19 and 21 inches. These diagonal measurements, unfortunately, represent not the actual screen that will be displayed but the size of the tube. As a result, comparing one company's 15-inch monitor to that of another may be unfair unless you actually measure the active screen area. This area can vary slightly from monitor to monitor, so one company's 17-inch monitor may display a 15.0-inch image, and another company's 17-inch monitor may present a 15.5-inch image.

The following table shows the advertised monitor diagonal size along with the approximate diagonal measure of the actual active viewing area for the most common display sizes:

The size of the actual viewable area varies slightly from manufacturer to manufacturer, but these figures are representative of most monitors. As you can see, the viewable area of a monitor is normally about 1.5 inches less than the advertised specification. The viewing area refers to the diagonal measure of the lighted area on the screen. In other words, if you are running Windows, the viewing area is the actual diagonal measure of the desktop.

In most cases, a 17-inch monitor is recommended as the minimum you should consider for most normal applications. Low-end applications can still get away with a 15-inch display, but resolution will suffer. 18- to 21-inch or larger displays are recommended for high-end systems, especially where graphics applications are the major focus.

Larger monitors are handy for applications such as desktop publishing, in which the smallest details must be clearly visible. With a 17-inch or larger display, you can see nearly an entire 8 1/2x11-inch page in 100 percent view--in other words, what you see on-screen virtually matches the page that will be printed. This feature is called WYSIWYG--short for "What You See Is What You Get." If you can see the entire page at its actual size, you can save yourself the trouble of printing several drafts before you get it right.

With the popularity of the Internet, monitor size and resolution became even more of an issue. Many Web pages are designed for 1,024x768 resolution, which requires a 17-inch CRT display as a minimum to handle without eye strain and inadequate focus. Because of their much tighter dot pitch, LCD displays in laptop computers can handle that resolution easily on 13.3- or even 12.1-inch displays. Using 1,024x768 resolution means you will be able to view most Web pages without scrolling sideways, which is a major convenience.

Monitor Resolution

Resolution is the amount of detail that a monitor can render. This quantity is expressed in the number of horizontal and vertical picture elements, or pixels, contained in the screen. The greater the number of pixels, the more detailed the images. The resolution required depends on the application. Character-based applications (such as word processing) require little resolution, whereas graphics-intensive applications (such as desktop publishing and Windows software) require a great deal.

There are several standard resolutions available in PC graphics adapters. The following table lists the standard resolutions used in PC video adapters and the term used to commonly describe them:

In a monochrome monitor, the picture element is a screen phosphor, but in a color monitor, the picture element is a phosphor triad. This difference raises another consideration called dot pitch, which applies only to color monitors. Dot pitch is the distance, in millimeters, between phosphor triads. Screens with a small dot pitch contain less distance between the phosphor triads; as a result, the picture elements are closer together, producing a sharper picture. Conversely, screens with a large dot pitch tend to produce images that are less clear.

The original IBM PC color monitor had a dot pitch of 0.43mm, which is considered to be poor by almost any standard. A dot pitch of more than 0.28mm is not recommended in most cases. While you can save money by picking a smaller monitor or one with a higher dot pitch, the trade-off is not usually worth it.

Interlaced versus Noninterlaced

Monitors and video adapters may support interlaced or noninterlaced resolution. In noninterlaced (conventional) mode, the electron beam sweeps the screen in lines from top to bottom, one line after the other, completing the screen in one pass. In interlaced mode, the electron beam also sweeps the screen from top to bottom, but it does so in two passes--sweeping the odd lines first and the even lines second. This mode is sometimes activated to use high resolutions on a small monitor, which does not support the higher refresh rate together with the higher resolution.

The drawback is that interlacing depends on the ability of the eye to combine two nearly identical lines, separated by a gap, into one solid line. On most monitors, interlacing produces a pronounced flicker in the display, unless the phosphor is designed with a very long persistence. If you are looking for high-quality video, however, you want to get a video adapter and monitor that support high-resolution, noninterlaced displays.

Energy and Safety

A properly selected monitor can save energy. The Environmental Protectal Agency has designed a standard called Energy Star. Any PC-and-monitor combination that consumes less than 60 watts (30 watts apiece) during idle periods can use the Energy Star logo. Some research shows that such "green" PCs can save a user a lot of money in electricity costs.

Monitors, being one of the most power-hungry computer components, can contribute to those savings. Perhaps the best-known energy-saving standard for monitors is VESA's Display Power-Management Signaling (DPMS) spec, which defines the signals that a computer sends to a monitor to indicate idle times. The computer or video card decides when to send these signals.

If you buy a DPMS monitor, you can take advantage of energy savings without remodeling your entire system. If you do not have a DPMS-compatible video adapter, some cards can be upgraded to DPMS with a software utility. Similarly, some energy-saving monitors include software that works with almost any graphics card to supply DPMS signals.

Another trend in green monitor design is to minimize the user's exposure to potentially harmful electromagnetic fields. Several medical studies indicate that these electromagnetic emissions may cause health problems, such as miscarriages, birth defects, and cancer. The risk may be low, but if you spend a third of your day (or more) in front of a computer monitor, that risk is increased.

The concern is that VLF (very low frequency) and ELF (extremely low frequency) emissions might affect the body. These two emissions come in two forms: electric and magnetic. Some research indicates that ELF magnetic emissions are more threatening than VLF emissions, because they interact with the natural electric activity of body cells. Monitors are not the only culprits; significant ELF emissions also come from electric blankets and power lines.

---

NOTE: ELF and VLF are a form of electromagnetic radiation; they consist of radio frequencies below those used for normal radio broadcasting.

---

These two frequencies are covered by the Swedish monitor-emission standard called SWEDAC, named after the Swedish regulatory agency. In many European countries, government agencies and businesses buy only low-emission monitors. The degree to which emissions are reduced varies from monitor to monitor. The Swedish government's MPR I standard, which dates back to 1987, is the least restrictive. MPR II, established in 1990, is significantly stronger (adding maximums for ELF as well as VLF emissions).

A more stringent 1992 standard called TCO further tightens the MPR II requirements. In addition, it is a more broad-based environmental standard that includes power-saving requirements and emission limits.

If you do not have a low-emission monitor, you can take other steps to protect yourself. The most important is to stay at arm's length (about 28 inches) from the front of your monitor. When you move a couple of feet away, ELF magnetic emission levels usually drop to those of a typical office with fluorescent lights. Likewise, monitor emissions are weakest at the front of a monitor, so stay at least 3 feet from the sides and backs of nearby monitors and 5 feet from any photocopiers, which are also strong sources of ELF.

Electromagnetic emissions should not be your only concern; you also should be concerned about screen glare. In fact, some antiglare screens not only reduce eye strain but also cut ELF and VLF emissions.

Monitor Buying Criteria

A monitor may account for as much as 50 percent of the price of a computer system. What should you look for when you shop for a monitor?

The trick is to pick a monitor that works with your selected video card. You can save money by purchasing a single-standard (fixed-frequency) monitor and a matching video card; for example, you can order a VGA monitor and a VGA video card. For greatest flexibility, get a multisync monitor that accommodates a range of standards.

With multisync monitors, you must match the range of horizontal and vertical frequencies the monitor accepts with those generated by your video card. The wider the range of signals, the more expensive--and more versatile--the monitor. Your video card's vertical and horizontal frequencies must fall within the ranges supported by your monitor. The vertical frequency (or refresh rate) determines how stable your image will be. The higher the vertical frequency, the better. Typical vertical frequencies range from 50 to 100Hz or more. The horizontal frequency (or line rate) ranges between 31.5KHz to 120KHz or more.

To keep the horizontal frequency low, some video cards use interlaced signals, alternately displaying half the lines of the total image. On most monitors, interlacing produces a pronounced flicker in the display, unless the phosphor is designed with a very long persistence. For this reason, you should avoid using interlaced video modes if possible. Some older cards and displays used interlacing as an inexpensive way to attain a higher resolution than otherwise would be possible. For example, the original IBM XGA adapters and monitors used an interlaced vertical frame rate of 43.5Hz in 1,024x768 mode, instead of the 60Hz or higher frame rate that most other adapters and displays use at that resolution.

A 60Hz vertical scan frequency (refresh or frame rate) is the minimum anybody should use, and even at this frequency a flicker will be noticed by most people. Especially on a larger display, this can cause eye strain and fatigue. If you can select a frame rate of 72Hz or higher, most people will not be able to discern any flicker. Most displays easily handle vertical frequencies of 85Hz or more, which greatly reduces the flicker seen by the user. Note that increasing the frame rate can slow down the video hardware, because it now needs to display each image more times per second.

When you shop for a VGA monitor, make sure that the monitor supports a horizontal frequency of at least 31.5KHz--the minimum that a VGA card needs to paint a 640x480 screen. The VESA Super VGA (800x600) or SVGA standard requires a 72Hz vertical frequency and a horizontal frequency of at least 48KHz. The sharper 1,024x768 image requires a vertical frequency of 60Hz and a horizontal frequency of 58KHz. For a super-crisp display, look for available vertical frequencies of 75Hz or higher and horizontal frequencies of up to 90KHz or more.

Suffice it to say that before investing in a monitor, you should check the technical specifications to make sure that the monitor meets your needs. If you are looking for a place to start, check out some of the different magazines, which periodically feature reviews of monitors.

Many inexpensive monitors are curved because it is easier to send an electron beam across them. Flat-screen monitors, which are a bit more expensive, look better to most people. As a general rule, the less curvature a monitor has, the less glare it will reflect.

Consider the size of your desk before you think about a monitor of 17 inches or larger. A 17-inch monitor typically is at least 1 1/2 feet deep, and a 20-inch monitor takes up 2 square feet. Typical 14-inch monitors are 16 to 18 inches deep.

You also should check the dot pitch of the monitor. Smaller pitch values indicate sharper images. Be wary of monitors with anything larger than a 0.28mm dot pitch; they lack clarity for fine text and graphics.

What resolution do you want for your display? Generally, the higher the resolution, the larger the display you will want. If you are operating at 640x480 resolution, for example, you should find a 15-inch monitor to be comfortable. At 1,024x768, you probably will find that the display of a 15-inch monitor is too small and therefore will prefer to use a larger one, such as a 17-inch monitor.

Here are the recommended minimum monitor sizes to properly display popular VGA and SVGA resolutions:

The minimum recommended display size is the advertised diagonal display dimension of the monitor. Note that this is not what the monitor may be capable of, but is what is recommended. In other words, most 15-inch monitors will display resolutions at least up to 1,024x768, but the characters, icons, and displayed information will be too small and will cause eye strain if you try to run beyond the 800x600 recommended. In other words, if you plan on spending a lot of time in front of your PC, and you want to run 1,024x768 resolution, a 17-inch display is absolutely recommended. Anything smaller is not considered proper ergonomics, and eye strain, headaches, and fatigue can result.

One exception to this rule is with the laptop and notebook displays. These are usually an LCD-type display, which is always crisp and perfectly focused by nature. Also, the dimensions advertised for the LCD screens are exactly what you get for display, unlike conventional CRT-based monitors. So the 12.1-inch LCD panels found on many laptop systems actually have a viewable area that is 12.1-inch diagonal. In other words, 12.1-inch is the size of the Windows desktop or functional area of the screen. This measurement compares to a 14-inch or even 15-inch conventional display in most cases. Not only that, but the LCD is so crisp that you can easily handle resolutions that are higher than otherwise would be acceptable on a CRT. For example, many laptop systems use 13.3-inch LCD panels that feature 1,024x768 resolution. Although this resolution is unacceptable on a 14-inch or 15-inch CRT display, it works well on the 13.3-inch LCD panel due to the crystal clear image.

---

TIP: Get a monitor with positioning and image controls that are easy to reach. A tilt-swivel stand should be included with your monitor, enabling you to move the monitor to the best angle for your use.

---

Most of the newer monitors use digital controls instead of analog controls. This has nothing to do with the signals sent to the monitor, but the controls (or lack of them) on the front panel. Monitors with digital controls have a built-in menu system that allows you to set things like brightness, contrast, screen size, vertical and horizontal shifts, and even focus. The menu is brought up on the screen by a button, and you use controls to make menu selections and vary the settings. When completed, the monitor saves your settings in NVRAM (Non-Volatile RAM) in the monitor. These settings are permanently stored using no battery, and can be altered at any time in the future. Most monitors can save different settings for different frequencies. While switching from one resolution to another, the monitor jumps to the settings you previously saved using that resolution. Digital controls give a much higher level of control over the monitor, and are highly recommended.

A monitor is such an important part of your computer that it is not enough to know just its technical specifications. Knowing a monitor has a 0.28mm dot pitch does not necessarily tell you that it is ideal for you. It is best to "kick the tires" of your new monitor at a showroom or (with a liberal return policy) in the privacy of your office. To test your monitor:

• Draw a circle with a graphics program. If the result is an oval, not a circle, this monitor will not serve you well with graphics or design software.
  
  
• Type some words in 8- or 10-point type (1 point equals 1/72 inch). If the words are fuzzy, or if the black characters are fringed with color, select another monitor.
  
  
• Turn the brightness up and down while examining the corner of the screen's image. If the image blooms or swells, it is likely to lose focus at high brightness levels.
  
  
• Load Microsoft Windows to check for uniform focus. Are the corner icons as sharp as the rest of the screen? Are the lines in the title bar curved or wavy? Monitors usually are sharply focused at the center, but seriously blurred corners indicate a poor design. Bowed lines may be the result of a poor graphics card, so don't dismiss a monitor that shows those lines without using another card to double-check the effect.
  
  
• A good monitor will be calibrated so that rays of red, green, and blue light hit their targets (individual phosphor dots) precisely. If they don't, you have bad convergence. This is apparent when edges of lines appear to illuminate with a specific color. If you have good convergence, the colors will be crisp, clear, and true, provided that there is not a predominant tint in the phosphor.

Video Cards

A video card provides signals that operate your monitor. With the PS/2 systems introduced in 1987, IBM developed new video standards that have overtaken the older display standards in popularity and support.

Most video cards follow one of several industry standards:

These adapters and video standards are supported by virtually every program that runs on IBM or compatible equipment. Most microcomputer monitors support at least one video standard, enabling you to operate them with video cards and software that are compatible with that standard. For example, a monitor that supports VGA may operate with VGA video cards and VGA software.

Obsolete Display Adapters

Although many types of display systems are considered to be standards, not all systems are considered to be viable standards for newer hardware and software. For example, the CGA standard works but is unacceptable for running the graphics-intensive programs on which many users rely. In fact, Microsoft Windows 3.1 does not work with any PC that has less-than-EGA resolution, and Windows 95 and Windows NT require VGA as an absolute minimum. The next several sections discuss the display adapters that are viewed as being obsolete.

Monochrome Display Adapter (MDA) and Display

The simplest (and first available) display type is the IBM Monochrome Display Adapter (MDA). It was introduced along with the IBM PC itself in 1981. The MDA video card can display text only at a 720x350 resolution. One interesting point is that the MDA card also incorporated a printer port and was the first multi-function adapter card available. It is a character-only system, the display has no inherent graphics capabilities. The display originally was a top-selling option because it is fairly cost-effective. As a bonus, the MDA provides a printer interface, conserving an expansion slot.

The display was known for clarity and high resolution, making it ideal for business use--especially for businesses that used DOS-based word processing or spreadsheet programs. Figure 10.1 shows the Monochrome Display Adapter.

FIG. 10.1  The Monochrome Display Adapter and connector.

The MDA uses a 9-pin D-shell connector. Table 10.1 shows the pinouts of the MDA connector.

Table 10.1  The IBM Monochrome Display Adapter (MDA) Connector

Because the monochrome display is a character-only display, you cannot use it with software that requires graphics. Originally, that drawback only kept the user from playing games on a monochrome display. With the 9x14 dot character box (matrix), the IBM monochrome monitor displays attractive characters.

Table 10.2 summarizes the features of the MDA's single mode of operation.

Table 10.2  IBM Monochrome Display Adapter (MDA) Specifications

Later, a company named Hercules Computer Technology Inc. released a video card called the Hercules Graphics Card (HGC). This card displays sharper text and can handle graphics, such as bar charts and graphic games. The resolution of the Hercules Graphics Card is 720x348, and is compatible with the IBM Monochrome Display Adapter.

Color Graphics Adapter (CGA) and Display

The Color Graphics Adapter (CGA) was introduced along with the IBM PC itself in 1981 and for many years was the most common video card. This adapter has two basic modes of operation: alphanumeric (A/N) or all points addressable (APA). In A/N mode, the card operates in 40-column by 25-line mode or 80-column by 25-line mode with 16 colors. In APA and A/N modes, the character set is formed with a resolution of 8x8 pixels. In APA mode, two resolutions are available: medium-resolution color mode (320x200), with four colors available from a palette of 16; and two-color high-resolution mode (640x200). Figure 10.2 shows the Color Graphics Adapter.

FIG. 10.2  The Color Graphics Adapter and connector.

The CGA uses a 9-pin D-shell connector. Table 10.3 shows the pinouts of the CGA connector.

Table 10.3  The Color Graphics Adapter (CGA) Connector

Also a composite video interface is part of the specification. The composite interface uses a standard pin plug or RCA plug. Most of the monitors sold for the CGA are RGBs, not composite monitors. The color signal of a composite monitor contains a mixture of colors that must be decoded or separated. RGB monitors receive red, green, and blue separately, and combine the colors in different proportions to generate other colors. RGB monitors offer better resolution than composite monitors, and they do a much better job of displaying 80-column text.

Also an RF modulator interface and a light-pen interface are part of the specification. Figure 10.3 shows the location of the connectors for the RF modulator and the light-pen interface.

FIG. 10.3  CGA RF modulator and light-pen connectors.

The RF modulator interface uses a 4-pin-header connector. Table 10.4 shows the pinouts of the RF modulator interface.

Table 10.4  The RF Modulator Interface Connector

The light-pen interface uses a 6-pin-header connector. Table 10.5 shows the pinouts of the light-pen interface.

Table 10.5  The Light-Pen Interface Connector

One drawback of a CGA video card is the fact that it produces flicker and snow. Flicker is the annoying tendency of the text to flash as you move the image up or down. Snow is the flurry of bright dots that can appear anywhere on the screen.

Table 10.6 lists the specifications for all CGA modes of operation.

Table 10.6  IBM Color Graphics Adapter (CGA) Specifications

APA = All points addressable (graphics)
-- = Not supported

Enhanced Graphics Adapter (EGA) and Display

The IBM Enhanced Graphics Adapter was introduced in 1984, just after the IBM AT system. It was discontinued when the PS/2 systems were introduced in April 1987. It consists of a graphics board, a graphics memory-expansion board, a graphics memory-module kit, and a high-resolution color monitor. The whole package originally cost about $1,800! The aftermarket gave IBM a great deal of competition in this area; it was possible to put together a similar system from non-IBM vendors for much less money. One advantage of EGA, however, is that you can build your system in modular steps. Because the card works with any of the monitors IBM produced at the time, you can use it with the IBM Monochrome Display, the earlier IBM Color Display, or the IBM Enhanced Color Display.

With the EGA card, the IBM color monitor displays 16 colors in 320x200 or 640x200 mode, and the IBM monochrome monitor shows a resolution of 640x350 with a 9x14 character box (text mode). Figure 10.4 shows the Enhanced Graphics Adapter.

FIG. 10.4  The Enhanced Graphics Adapter and connector.

The EGA uses a 9-pin D-shell connector. Table 10.7 shows the pinouts of the EGA connector.

Table 10.7  The Enhanced Graphics Adapter (EGA) Connector

Also a light-pen interface is part of the specification. Figure 10.5 shows the location of the connector for the light-pen interface.

FIG. 10.5  EGA light-pen connector.

The light-pen interface uses a 6-pin-header connector. Table 10.8 shows the pinouts of the light-pen interface.

Table 10.8  The Light-Pen Interface Connector

With the EGA card, the IBM Enhanced Color Display is capable of displaying 640x350 pixels in 16 colors from a palette of 64. The character box for text is 8x14, compared with 8x8 for the earlier CGA board and monitor. The 8x8 character box can be used, however, to display 43 lines of text. Through software, the character box can be manipulated up to the size of 8x32.

You can enlarge a RAM-resident, 256-member character set to 512 characters by using the IBM memory expansion card. A 1,024-character set is added with the IBM graphics memory-module kit. These character sets are loaded from programs.

All this memory fits in the unused space between the end of RAM user memory and the current display-adapter memory. The EGA has a maximum 128K of memory that maps into the RAM space just above the 640K boundary. If you install more than 640K, you will probably lose the extra memory after installing the EGA. The graphics memory-expansion card adds 64K to the standard 64K, for a total 128K. The IBM graphics memory-module kit adds another 128K, for a total 256K. This second 128K of memory is only on the card and does not consume any of the PC's memory space. (Because almost every aftermarket EGA card comes configured with the full 256K of memory, expansion options are not necessary.)

Table 10.9 shows the modes supported by the EGA adapter.

Table 10.9  IBM Enhanced Graphics Adapter (EGA) Specifications

APA = All points addressable (graphics)

The EGA system has some problems emulating the earlier CGA or MDA adapters, and some software that works with the earlier cards will not run on the EGA until the programs are modified.

Professional Color Display and Adapter

The Professional Graphics Display System is a video display product that IBM introduced in 1984. At $4,290, the system was too expensive to become a mainstream product and never achieved any popularity. It was the first processor-based video adapter for PCs; it actually incorporated an 8088 processor on the card itself.

The system consists of a Professional Graphics Monitor and a Professional Graphics Card Set. When fully expanded, this card set used three slots in an XT or AT system--a high price to pay, but the features were impressive. The Professional Graphics Adapter (PGA) offers three-dimensional rotation and clipping as a built-in hardware function. The adapter can run 60 frames of animation per second because the PGA uses a built-in dedicated microprocessor. Figure 10.6 shows the Professional Graphics Adapter.

FIG. 10.6  The Professional Graphics Adapter (PGA) and connector.

The PGA uses a 9-pin D-shell connector. Table 10.10 shows the pinouts of the PGA connector.

Table 10.10  The Professional Graphics Adapter (PGA) Connector

* This signal was very rarely implemented. Most PGA equipment either does not use pin 5 or leaves it unconnected.

The Professional Graphics card and monitor targeted engineering and scientific applications rather than financial or business applications. This system, which was discontinued when the PS/2 was introduced, has been replaced by the VGA and other higher-resolution graphics standards for these newer systems.

Table 10.11 lists all supported PGA modes.

Table 10.11  IBM Professional Graphics Adapter (PGA) Specifications

APA = All points addressable (graphics)
-- = Not supported

8514/A Display Adapter

The PS/2 8514/A Display Adapter, introduced in 1987 along with the PS/2 systems, offers higher resolution and more colors than the standard VGA. This adapter, designed to use the PS/2 Color Display 8514, plugs into a Micro Channel slot in any PS/2 model so equipped.

All operation modes of the built-in VGA continue to be available. An IBM 8514 memory-expansion kit is available for the 8514/A. This kit provides increased color and grayscale support.

To take full advantage of this adapter, you should use the 8514 display because it is matched to the capabilities of the adapter. Notice that IBM later discontinued the 8514/A adapter and specified the XGA in its place.

The 8514/A standard uses the same 15-pin High Density D-Shell connector as the VGA standard. See the section "VGA Adapters and Displays" later in this chapter for the pinouts of this connector. Table 10.12 shows all 8514/A modes.

Table 10.12  IBM 8514/A Display Adapter Specifications

APA = All points addressable (graphics)
* = Interlaced

MultiColor Graphics Array (MCGA)

The MultiColor Graphics Array (MCGA) is a graphics adapter that is integrated into the motherboard of the PS/2 Models 25 and 30. The MCGA supports all CGA modes when an IBM analog display is attached, but any previous IBM display is not compatible. In addition to providing existing CGA mode support, the MCGA includes four additional modes.

The MCGA uses as many as 64 shades of gray in converting color modes for display on monochrome monitors, so that users who prefer a monochrome display still can execute color-based applications.

The MCGA standard uses the same 15-pin High Density D-Shell connector as the VGA standard. See the section "VGA Adapters and Displays" later in this chapter for the pinouts of this connector. Table 10.13 lists the MCGA display modes.

Table 10.13  IBM MultiColor Graphics Array (MCGA) Specifications

VGA Adapters and Displays

When IBM introduced the PS/2 systems on April 2, 1987, it also introduced the Video Graphics Array (VGA) display. On that day, in fact, IBM also introduced the lower-resolution MultiColor Graphics Array (MCGA) and higher-resolution 8514 adapters. The MCGA and 8514 adapters did not become popular standards like the VGA, and both were discontinued very soon.

Digital Versus Analog Signals

Unlike earlier video standards, which are digital, the VGA is an analog system. Why are displays going from digital to analog when most other electronic systems are going digital? Compact-disc players (digital) have replaced the turntables (analog), and newer VCRs and camcorders have digital picture storage for smooth slow-motion and freeze-frame capability. With a digital television set, you can watch several channels on a single screen by splitting the screen or placing a picture within another picture.

Why, then, did IBM decide to change the video to analog? The answer is color.

Most personal-computer displays introduced before the PS/2 are digital. This type of display generates different colors by firing the RGB electron beams in on-or-off mode. You can display up to eight colors (2 to the third power). In the IBM displays and adapters, another signal intensity doubles the number of color combinations from 8 to 16 by displaying each color at one of two intensity levels. This digital display is easy to manufacture and offers simplicity with consistent color combinations from system to system. The real drawback of the digital display system is the limited number of possible colors.

In the PS/2 systems, IBM went to an analog display circuit. Analog displays work like the digital displays that use RGB electron beams to construct various colors, but each color in the analog display system can be displayed at varying levels of intensity--64 levels, in the case of the VGA. This versatility provides 262,144 possible colors (64³). For realistic computer graphics, color often is more important than high resolution, because the human eye perceives a picture that has more colors as being more realistic. IBM moved graphics into analog form to enhance the color capabilities.

Video Graphics Array (VGA)

PS/2 systems contain the primary display adapter circuits on the motherboard. The circuits, or VGA, are implemented by a single custom VLSI chip designed and manufactured by IBM. To adapt this new graphics standard to the earlier systems, IBM introduced the PS/2 Display Adapter. Also called a VGA card, this adapter contains the complete VGA circuit on a full-length adapter board with an 8-bit interface.

The VGA BIOS (Basic Input/Output System) is the control software residing in the system ROM for controlling VGA circuits. With the BIOS, software can initiate commands and functions without having to manipulate the VGA directly. Programs become somewhat hardware-independent and can call a consistent set of commands and functions built into the system's ROM-control software.

Newer implementations of the VGA can be different in hardware, but they respond to the same BIOS calls and functions. The VGA, therefore, is compatible with the graphics and text BIOS functions that were built into the PC systems from the beginning. The VGA can run almost any software that originally was written for the MDA, CGA, or EGA.

In a perfect world, software programmers would write to the BIOS interface rather than directly to the hardware and would promote software interchanges between different types of hardware. More frequently, however, programmers want the software to perform better, so they write the programs to control the hardware directly. As a result, these programmers achieve higher-performance applications that are dependent on the hardware for which they were first written.

When bypassing the BIOS, a programmer must ensure that the hardware is 100 percent compatible with the standard so that software written to a standard piece of hardware runs on the system. Just because a manufacturer claims this register level of compatibility does not mean that the product is 100 percent compatible or that all software runs as it would on a true IBM VGA. Most manufacturers have "cloned" the VGA system at the register level, which means that even applications that write directly to the video registers will function correctly. Also, the VGA circuits themselves emulate the older adapters even to the register level and have an amazing level of compatibility with these earlier standards. This compatibility makes the VGA a truly universal standard.

The VGA displays up to 256 colors on screen, from a palette of 262,144 (256K) colors. Because the VGA outputs an analog signal, you must have a monitor that accepts an analog input.

VGA displays come not only in color but also in monochrome VGA models, using color summing. With color summing, 64 gray shades are displayed instead of colors; the translation is performed in the ROM BIOS. The summing routine is initiated if the BIOS detects the monochrome display when the system is booted. This routine uses a formula that takes the desired color and rewrites the formula to involve all three color guns, producing varying intensities of gray. The color that would be displayed, for example, is converted to 30 percent red plus 59 percent green plus 11 percent blue to achieve the desired gray. Users who prefer a monochrome display, therefore, can execute color-based applications.

Table 10.14 lists the VGA display modes.

Table 10.14  IBM Video Graphics Array (VGA) Specifications

XGA and XGA-2

IBM announced the PS/2 XGA Display Adapter/A on October 30, 1990, and the XGA-2 in September 1992. Both adapters are high-performance, 32-bit bus-master adapters for Micro Channel-based systems. These video subsystems, evolved from the VGA, provide greater resolution, more colors, and much better performance. Combine fast VGA, more colors, higher resolution, a graphics coprocessor, and bus-mastering, and you have XGA. Being a bus-master adapter means that the XGA can take control of the system as though it were the motherboard. In essence, a bus master is an adapter with its own processor that can execute operations independent of the motherboard.

The XGA was introduced as the default graphics-display platform with the Model 90 XP 486 and the Model 95 XP 486. In the desktop Model 90, the XGA is on the motherboard; in the Model 95 (a tower unit), it is located on a separate add-in board. This board--the XGA Display Adapter/A--also is available for other 386- and 486-based Micro Channel systems. The XGA adapter can be installed in any MCA systems that have 80386SX or faster processors.

The XGA comes standard with 512K of graphics memory, which can be upgraded to 1M with an optional video-memory expansion.

In addition to all VGA modes, the XGA adapter offers several new modes of operation, which are listed in Table 10.15.

Table 10.15  XGA Unique Modes of Operation

The reasons for the different memory requirements are explained in the next section. The 65,536-color mode provides almost photographic output. The 16-bit pixel is laid out as 5 bits of red, 6 bits of green, and 5 bits of blue (5-6-5)--in other words, 32 shades of blue, 64 shades of green, and 32 shades of blue. (The eye notices more variations in green than in red or blue.) One major drawback of the XGA implementation is the interlacing that occurs in the higher-resolution modes. With interlacing, you can use a less expensive monitor, but the display updates more slowly, resulting in a slight flicker.

The XGA-2 improves on the performance of the XGA in several ways. To begin with, the XGA-2 increases the number of colors supported at 1,024x768 resolution to 64K. In addition, because of the circuitry of the XGA-2, it can process data at twice the speed of the XGA. The XGA-2 also works in noninterlaced mode, so it produces less flicker than the XGA does.

Both the XGA and XGA-2 support all existing VGA and 8514/A video modes. A large number of popular applications have been developed to support the 8514/A high-resolution 1,024x768 mode. These applications are written to the 8514/A Adapter interface, which is a software interface between the application and the 8514/A hardware. The XGA's extended graphics function maintains compatibility at the same level. Because of the power of the XGA and XGA-2, existing VGA or 8514/A applications run much faster.

Much of the speed of the XGA and XGA-2 also can be attributed to its Video RAM (VRAM), a type of dual-ported RAM designed for graphics-display systems. This memory can be accessed by both the processor on the graphics adapter and the system CPU simultaneously, providing almost instant data transfer. The XGA VRAM is mapped into the system's address space. The VRAM normally is located in the top addresses of the 386's 4G address space. Because no other cards normally use this area, conflicts are rare. The adapters also have an 8K ROM BIOS extension that must be mapped somewhere in segments C000 or D000. (The motherboard implementation of the XGA does not require its own ROM, because the motherboard BIOS contains all the necessary code.)

Table 10.16 summarizes the XGA modes.

Table 10.16  IBM eXtended Graphics Array (XGA) Specifications

Super VGA (SVGA)

When IBM's XGA and 8514/A video cards were introduced, competing manufacturers chose not to clone these incremental improvements on VGA. Instead, they began producing lower-cost adapters that offered even higher resolutions. These video cards fall into a category loosely known as Super VGA (SVGA).

SVGA provides capabilities that surpass those offered by the VGA adapter. Unlike the display adapters discussed so far, SVGA refers not to a card that meets a particular specification but to a group of cards that have different capabilities.

For example, one card may offer several resolutions (such as 800x600 and 1,024x768) that are greater than those achieved with a regular VGA, whereas another card may offer the same or even greater resolutions but also provide more color choices at each resolution. These cards have different capabilities; nonetheless, both are classified as SVGA.

The SVGA cards look much like their VGA, 8514/A, MCGA and XGA counterparts. They all use the same 15-pin High Density D-Shell connector. Figure 10.7 shows the Video Graphics Array adapter.

FIG. 10.7  The Video Graphics Array (VGA) adapter and connector.

Because the technical specifications from different SVGA vendors vary tremendously, it is impossible to provide a definitive technical overview in this documentation. The VGA, 8514/A, MCGA, XGA and SVGA standards all use the same connector and pinouts. Table 10.17 shows the pinouts for this connector.

Table 10.17  The VGA, 8514/A, MCGA, XGA and SVGA Connector

* Pin 5 is sometimes used as a Ground, and sometimes it carries a Self-Test signal.
** In the original VGA standard, if the adapter senses that pin 11 is a Ground and pin 12 is not connected, it assumes a color monitor and outputs color VGA. If it senses that pin 11 is not connected and pin 12 is a Ground, it assumes a monochrome monitor and outputs monochrome VGA.

In some cases, a 9-pin D-Shell connector is used voor VGA or SVGA. The pinouts for this de facto standard are shown in Table 10.18.

Table 10.18  The 9-pin VGA/SVGA Connector

The Video Electronics Standards Association (VESA) includes members from various companies associated with PC and computer video products. In October 1989, recognizing that programming for the many SVGA cards on the market was virtually impossible, VESA proposed a standard for a uniform programmer's interface for SVGA cards.

The SVGA standard is called the VESA BIOS Extension. If a video card incorporates this standard, a program easily can determine the capabilities of the card and access them. The benefit of the VESA BIOS Extension is that a programmer needs to worry about only one routine or driver to support SVGA. Different cards from different manufacturers are accessible through the common VESA interface.

When first proposed, this concept met with limited acceptance. Several major SVGA manufacturers started supplying the VESA BIOS Extension as a separate memory-resident program that you could load when you booted your computer. Over the years, however, other vendors started supplying the VESA BIOS Extension as an integral part of their SVGA BIOS. Obviously, from a user's perspective, support for VESA in BIOS is a better solution. You do not have to worry about loading a driver or other memory-resident program whenever you want to use a program that expects the VESA extensions to be present.

Even if a SVGA video adapter claims to be VESA-compatible, however, it still may not work with a particular driver, such as the 800x600, 256-color, SVGA driver that comes with Microsoft Windows. In practice, however, manufacturers continue to provide their own driver software.

Table 10.19 lists the video modes of the Chips and Technologies 65554 SVGA graphics accelerator, a typical and often used chipset.

Table 10.19  Chips and Technologies 65554 Graphics Accelerator Chipset Video Modes

The VESA Feature Connector (VFC)

The VESA Feature Connector (VFC) is an additional connector that is used to connect the video chipset to other video devices, such as 3D accelerators, MPEG decoders and video capture cards. The reason that these connectors are used is that they permit the direct transfer of video information from these devices to the video card, without having to use the main system bus. Even high-performance local buses can get slowed down when trying to deal with the enormous amount of information that, for example, a full-motion video stream represents. The VFC interface is 8-bit wide, resulting in a maximum resolution of 640x480 with a maximum of 256 colors (standard VGA). Table 10.20 shows the pinouts of the 26-pin-header VESA Feature Connector.

Table 10.20  VESA Feature Connector

The VESA Advanced Feature Connector (VAFC) is developed as an extension to the VESA Feature Connector. Version 1.0 of the VAFC was released in December 1995. The VAFC interface widens the port from 8 bits to 16 or 32, and provides improved signaling for more reliability. The maximum clock rate is 37.5MHz, resulting in a maximum data transfer rate of 150M/sec in 32-bit mode. An 80-pin connector is used. Table 10.21 shows the pinouts of the 80-pin VESA Advanced Feature Connector.

Table 10.21  VESA Advanced Feature Connector

The VESA Media Channel (VMC) is a new developed industry bus standard, dedicated specifically to the problem of transferring raw, uncompressed digital video and audio data through the PC, without incurring performance overheads. VMC operates at a clock speed of 33MHz, resulting in a maximum data transfer rate of 132M/sec. VMC supports 8-bit, 16-bit and 32-bit devices simultaneously, and handles up to 15 data streams at the time without affecting performance. The VESA Media Channel provides the option for a 68-pin multi-drop cable, allowing multiple devices to be combined in a modular fashion. Table 10.22 shows the pinouts of the 68-pin VESA Media Channel Connector.

Table 10.22  VESA Media Channel Connector

A video card relies on memory in drawing your screen. You can often select how much memory you want on your video card--for example, 256K, 512K, 1M, 2M, 4M or 8M. Adding more memory does not speed up your video card; instead, it enables the card to generate more colors and/or higher resolutions.

The amount of memory needed by a video adapter to display a particular resolution and color depth is a mathematical equation. There has to be a memory location used to display every dot (or pixel) on the screen, and the number of total dots is determined by the resolution. For example 1,024x768 resolution represents 786,432 dots on the screen. If you were to display that resolution with only two colors, you would only need 1 bit to represent each dot. If the bit were a 0, the dot would be black, and if it were a 1, the dot would be white. If you used 4 bits to control each dot, you could display 16 colors, since there are 16 combinations possible with a four-digit binary number (2 to the 4th power equals 16). If you multiplied the number of dots times the number of bits required to represent each dot, you have the amount of memory required to display that resolution. Here is how the calculation would work:

As you can see, to display only 16 colors at 1,024x768 resolution would require exactly 384K of RAM on the video card. Upping the color depth to 8 bits per pixel results in 256 possible colors, and a memory requirement of 786,432 bytes or 768K. Since no video card can install that exact amount, you would have to install an actual 1M on the video card.

In order to use the higher resolution modes and greater numbers of colors in SVGA cards, such cards will need more memory than the 256K found on a standard VGA adapter. Table 10.23 shows some of the requirements for SVGA cards based on resolution and color depth.

Table 10.23  Display Adapter Minimum Memory Requirements

From this table, you can see that a video adapter with 2M can display 65,536 colors in 1,024x768 resolution mode, but for a true color (16.8M colors) display, you would need to upgrade to 4M. A 24-bit (or true-color) video card can display photographic images by using 16.8 million colors. If you spend a lot of time working with graphics, you may want to invest in a 24-bit video card with up to 4M of RAM or more.

Another issue with respect to memory on the graphics adapter is how wide the access is between the graphics chipset and the memory on the adapter. The graphics chipset is usually a single large chip on the card that contains virtually all of the adapter's functions. It is wired directly to the memory on the card through a local bus. Most of the high-end adapters use an internal 64-bit or 128-bit wide memory bus. This jargon is confusing, because this does not refer to the kind of bus slot the card plugs into. In other words, when you read about a 64-bit graphics adapter, it is really a 32-bit (PCI or VLB) card that has a 64-bit local memory bus on the card itself.

Improving Video Speed

Many efforts have been made to improve the speed of video adapters because of the complexity and sheer data of the high-resolution displays used by newer software. The improvements in video speed are occurring along three fronts:

• Processor
  
  
• RAM
  
  
• Bus

The combination of these three is reducing the video bottleneck caused by the demands of graphical user interface software, such as Microsoft Windows.

The Video Processor

Three types of processors, or chipsets, can be used in creating a video card. The chipset used is, for the most part, independent of which video specification (VGA, SVGA, or XGA) the adapter follows.

The oldest technology used in creating a video adapter is known as frame-buffer technology. In this scheme, the video card is responsible for displaying individual frames of an image. Each frame is maintained by the video card, but the computing necessary to create the frame comes from the CPU of your computer. This arrangement places a heavy burden on the CPU, which could be busy doing other program-related computing.

At the other end of the spectrum is a chip technology known as coprocessing. In this scheme, the video card includes its own processor, which performs all video-related computations. This arrangement frees the main CPU to perform other tasks. Short of integrating video functions directly into the CPU, this chipset provides the fastest overall system throughput.

Between these two arrangements is a middle ground: a fixed-function accelerator chip. In this scheme the circuitry on the video card does many of the more time-consuming video tasks (such as drawing lines, circles, and other objects), but the main CPU still directs the card by passing graphics-primitive commands from applications, such as an instruction to draw a rectangle of a given size and color.

The Video RAM

Historically, most video adapters have used regular dynamic RAM (DRAM) to store video images. This type of RAM, although inexpensive, is rather slow. The slowness can be attributed to the need constantly to refresh the information contained within the RAM, as well as to the fact that DRAM cannot be read at the same time it is being written.

Most newer PC graphics cards need extremely high data transfer rates to and from the video memory. At a resolution of 1,024x768 and a standard refresh rate of 72Hz, the Digital to Analog Converter (DAC) on the card needs to read the contents of the video memory frame buffer 72 times per second. In true color (24-bits per pixel) mode, this means that the video memory must be read at the rate of about 170M/sec, which is just about the maximum rate available from a conventional DRAM design. Because of the high bandwidth required, a number of competing memory technologies have emerged to meet the performance needs of high-end video memory.

One of the newer memory designs used in video cards is EDO (Extended Data Out) RAM. EDO provides a wider effective bandwidth by offloading memory precharging to separate circuits, so that the next access can essentially begin before the last access has finished. As a result, EDO offers a 10 percent speed boost over DRAM, at a similar cost. EDO RAM was introduced by Micron Technologies. It was originally designed for use in main memories, but it is also used in video card applications. EDO chips are constructed using the same dies as conventional DRAM chips, and they differ from DRAMs only in how they are wired in final production. This method enables EDO chips to be made on the same manufacturing lines and at the same relative costs as DRAM.

VRAM (Video RAM) is a special type of memory that has been used in video cards for a long time. VRAM is designed to be dual-ported, which allows the processor or accelerator chip on the graphics card as well as the DAC or even the PC's own processor to access the RAM simultaneously. This allows for much greater performance than standard DRAM or even EDO, but it comes at a higher price.

WRAM, or Window RAM, is a modified VRAM-type dual-ported memory technology developed by Samsung that is aimed specifically at graphics cards. WRAM offers marginally better performance than standard VRAM at a lower cost. WRAM is used in many high-end graphics cards as a replacement for VRAM.

MDRAM (Multibank DRAM) is a newer type of memory that is explicitly aimed at graphics and video applications. Developed by MoSys Inc., MDRAMs are constructed of a large number of small (32K) banks. Traditionally, DRAM or VRAM is logically organized as a single, monolithic bank. Being organized into small banks allows MDRAMs to be installed in any size that is an integral multiple of 32K, instead of restricting the size to the traditional binary multiple sizes found in many video cards. This was a significant advantage for the cost-sensitive PC marketplace.

For example, a 1,024x768 true color (24-bit) graphics system requires 2.3M for the frame buffer plus some extra memory for off-screen storage. If 256Kx16 DRAMs and a 64-bit bus are used, the only workable memory size that accommodates this frame buffer is 4M, constructed of two banks of four chips each. However, with MDRAM, a memory system of 2.5M can be constructed of only two or three individual chips. This eliminates the waste of an extra 1.5M, and the total memory cost can be significantly reduced.

In addition to the better memory sizing, MDRAM organizes its internal banks off a narrow central bus, which allows access to each bank individually. As such, this design can complete a burst to or from one bank and then begin a burst to or from another, all in a single clock cycle, offering much higher performance than VRAM or WRAM.

SGRAM, or Synchronous Graphics RAM, is a high-end solution for very fast video card designs. This type of memory can operate at 66MHz or faster.

The Bus

You learned that certain video cards are designed for certain buses. For example, the VGA was designed for use with an MCA bus, just like the XGA and XGA-2. The bus system that you use in your computer affects the speed at which your system processes video information. The ISA bus offers a 16-bit data path at speeds of 8.33MHz. The EISA or MCA buses can process 32 bits of data at a time, but they also run at speeds up to 10MHz. (Don't confuse the bus speed with the CPU speed. For example, when the CPU runs at a speed of 100MHz, the bus still can handle only a limited level of speed.)

One improvement on this frontier was the VESA local bus (VL-Bus) standard. The VL-Bus standard typically is an addition to an existing bus technology. For example, you might have an ISA system that also contains a VL-Bus slot. Even if it is used in an ISA system, the VL-Bus processes 32 bits of data at a time and it operates at the full-rated speed of the CPU. By using a VL-Bus video card instead of an ISA video card, you can increase the video performance of a pc.

In June 1992, Intel Corporation introduced Peripheral Component Interconnect (PCI) as a blueprint for directly connecting microprocessors and support circuitry. PCI combines the speed of a local bus with microprocessor independence. PCI video cards, like VL-Bus video cards, can increase video performance dramatically. PCI video cards, by their design, are meant to be Plug and Play (PnP), meaning that they require little configuration.

VL-Bus and PCI have some important differences, as Table 10.24 shows.

Table 10.24  Local Bus Specifications

The Fastest Speed Possible

Fortunately, you can choose the best of each area--chipset, RAM, and bus--to achieve the fastest speed possible. The faster you want your card to perform, the more money you must spend.

The trick to choosing a video subsystem is making an early decision. As you research specifications for your entire system, you should pay attention to the video and make sure that it performs the way you want it to.

Video Card Buying Criteria

One trend is to display higher-resolution images on larger and larger monitors. The growth of multimedia also has encouraged users to invest in 24-bit video cards for photographic-quality images. Both of these trends mean that you may want your video card to produce its 16 million colors at very high resolutions. To avoid bothersome flickering images, make sure that your card supports at least 72Hz vertical refresh rates at all resolutions; 76Hz or 85Hz is even better.

Whichever card you buy, make sure that it also has on-board VGA support so that you do not need an extra VGA card. Drivers for your particular operating system should be included, as well as a utility for switching resolutions. Many of these utilities allow for video mode switching on-the-fly without having to leave the operating system environment. This capability for changing resolutions is standard in Windows 95 and was available for Windows 3.1 by using a utility program.

Chipsets

It is important to note that a video card essentially consists of four major components:

• Chipset
  
  
• DAC (Digital to Analog Converter)
  
  
• Video memory
  
  
• BIOS (Basic Input/Output System)

The chipset is the heart of any video card and essentially defines the card and its functions. Any two video cards built using the same chipset can have the same relative performance and capabilities. Also, when you install drivers, they are normally written by the chipset manufacturer and installed for your particular chipset rather than the card itself. Of course, cards built using the same chipset can differ in the amount and type of memory installed, so performance may vary.

When you inspect video cards for purchase, you should inquire about which chipset is used on the card; that will give you a much better basis for comparing that card against others. Also, knowing the chipset manufacturer of your card enables you to contact that manufacturer directly (via the internet, for example) so you can download the latest drivers in case of problems.

Video Cards for Multimedia

Multimedia is the result of several different media working together. Video is just one (important) element. Topics not yet discussed include animation, full-motion video (playback and capture), still images, and graphics processing. Still images and video provide dazzling slides, and animation and full-motion video breathe life into any presentation.

A computer can mathematically animate sequences between keyframes. A keyframe identifies specific points. A bouncing ball, for example, can have three keyframes: up, down, and up. Using these frames as a reference point, the computer can create all the images in between. This creates a smooth bouncing ball.

3-D graphics accelerator cards incorporate a chipset that is capable of on-board rendering. This enables smooth, photo-realistic 3-D images to be performed on a PC level at speeds exceeding those of low-end workstations.

Video Feature Connectors

Since IBM first developed the VGA standard in 1987, one often overlooked part of the standard was the VESA Feature Connector (VFC). This was a 26-pin connector that allowed other video cards to connect to a VGA adapter directly. Unfortunately, this standard was poorly documented by IBM and poorly implemented by most VGA adapter manufacturers. In fact, many VGA cards did not implement this connector at all, basically ignoring the need. That may have been fine in the early days of VGA, because there were few multimedia products that would need to tap into the VGA signal. Later, however, there came many types of multimedia add-on boards that had features such as motion video, video capture, television tuners, and so on that needed the services of this connector to do their job.

Unfortunately, there was another problem with the VFC besides it not being there or being implemented incorrectly. The problem was one of performance. The original VGA adapter was designed as an 8-bit bus adapter, and worked at a resolution of only 640x480 pixels. Thus, the VFC had these same limitations, which put a damper on the type of video signal that could be transferred directly from one card to another. Later, these problems were solved by release of the VESA Advanced Feature Connector (VAFC) and the VESA Media Channel (VMC) video bus standards. These new standards should bring about compatibility and performance for interconnected multimedia adapters and video adapters. These standards ensured rapid growth in the adoption of new applications such as interactive video, video presentation, video conferencing, and desktop video editing.

The VESA Advanced Feature Connector (VAFC) provides a low cost extension of the industry standard VFC found on many graphics boards. VAFC meets high bandwidth requirements by widening the current feature connector data path from 8 to 16/32-bits and adding additional signals, which provide more reliable operation. The VAFC delivers 75M/sec throughput in its 16-bit baseline configuration, and up to 150M/sec in the 32-bit configuration. Other features include multiple pixels per clock, color space data, genlocking, and asynchronous video input.

The VESA Media Channel (VMC) is a dedicated multimedia bus that provides an independent path for the simultaneous processing of several high bandwidth video streams. The VMC directly addresses the limitations of running video across a computer's system bus. This design solves the universal bandwidth bottleneck and latency issues that exist in most system or processor bus architectures including ISA, EISA, MicroChannel, VL-Bus, and PCI.

---

NOTE: Refer to the sections "The VESA Feature Connector (VFC)", "The VESA Advanced Feature Connector (VAFC)" and "The VESA Media Channel (VMC)" earlier in this chapter, for more information about the different types of Video Feature Connectors.

---

---

TIP: For any high performance video adapter, make sure that it supports at least the 80-pin VAFC connector or the 68-pin VMC connector. If you see only a 26-pin connector on the card, then the card would not be recommended as that is the standard VFC. Most of the higher-quality multimedia adapters will require a VAFC connection for high-performance video signal transfer.

---

Video Output Devices

When video technology was first introduced, it was based upon television. There is a difference between the signals a television uses and the signals used by a computer. In the United States, color TV standards were established in 1953 by the National Television System Committee (NTSC). Some countries, such as Japan, followed this standard. Many countries in Europe developed more sophisticated standards, including Phase Alternate Line (PAL) and SEquential Couleur Avec Memoire (SECAM). Table 10.25 shows the differences among these standards.

Table 10.25  Television versus Computer Monitors

A video-output (or VGA-to-NTSC/PAL/SECAM) adapter enables you to show computer screens on a TV set or record them onto videotape for easy distribution. These products fall into two categories: those with genlocking (which enables the board to synchronize signals from multiple video sources or video with PC graphics) and those without. Genlocking provides the signal stability needed to obtain adequate results when recording to tape but is not necessary for simply using a video display.

VGA-to-NTSC/PAL/SECAM converters come as both internal boards or external boxes that you can port along with your laptop-based presentation. These latter devices do not replace your VGA adapter but instead connect to your video adapter via an external cable that works with any type of VGA card. In addition to VGA input and output ports, a video-output board has a video output interface for S-Video and composite video.

The resolution shown on a TV set or recorded on videotape is often limited to straight VGA at 640x480 pixels. Such boards may contain an "anti-flicker" circuit to help stabilize the picture, which often suffers from a case of the jitters in VGA-to-TV products.

Still-Image Video Capture Cards

Like a Polaroid camera, you can capture individual screen images for later editing and playback. Some plug into a PC's parallel port. These units capture still images from NTSC/PAL/SECAM video sources like camcorders or VCRs. Although image quality is limited by the input signal, the results are still good enough for presentations and desktop publishing applications. These devices work with 8-, 16-, and 24-bit VGA cards and usually accept video input from VHS, Super VHS, and Hi-8 devices. As you might expect, though, Super VHS and Hi-8 video sources give better results, as do Super VGA modes with more than 256 colors.

You may want to invest in image-processing applications that offer features such as image editing, file conversion, screen capture, and graphics file management.

Desktop Video (DTV) Boards

You can also capture TV signals to your computer system for display or editing. When capturing video, you should think in terms of digital versus analog. The biggest convenience of an analog TV signal is efficiency; it is a compact way to transmit video information through a low-bandwidth pipeline. The disadvantage is that while you can control how the video is displayed, you cannot edit it.

Actually capturing and recording video from external sources and saving the files onto your PC requires special technology. What is needed is a video capture board, which is also referred to as a video digitizer or video grabber.

One of the most common uses for analog video is with interactive Computer-Based Training (CBT) programs in which your application sends start, stop, and search commands to a laserdisc player that plays disks you have mastered. The software controls the player via an interface that also converts the laserdisc's NTSC/PAL/SECAM signal into a VGA-compatible signal for display on your computer's monitor. These types of applications require TV-to-VGA conversion hardware.

Whereas a computer can display up to 16 million colors, the NTSC standard allows for only approximately 32,000 colors. Affordable video is the Achilles' heel of multimedia. The images are often jerky or less than full-screen. The reason is because full-motion video, such as you see on TV, requires 30 images or frames per second (fps).

The typical computer screen was designed to display mainly static images. The storing and retrieving of these images requires managing huge files. Consider this: A single, full-screen color image requires almost 2M of disk space; a one-second video would require 45M. Likewise, any video transmission that you want to capture for use on your PC must be converted from an analog NTSC signal to a digital signal that your computer can use. On top of that, the video signal must be moved inside your computer at 10 times the speed of the conventional ISA bus structure. You need not only a superior video card and monitor but also an excellent expansion bus, such as VL-Bus or PCI.

Considering the fact that full-motion video can consume massive quantities of disk space, it becomes apparent that compression is needed. Compression and decompression (or codec) applies to both video and audio. Not only does a compressed file take up less space, it also performs better; there is simply less data to process. When you are ready to replay the video/audio, you simply decompress the file during playback. In any case, ensure your hard drive is large enough and has enough performance to handle the huge files that can result in storing video capture files. A minimum of a 1 or 2G drive is recommended, with either an enhanced IDE or SCSI-2 interface.

There are two types of codecs: hardware-dependent codecs and software- (or hardware-independent) codecs. Hardware codecs are typically better; however, they require additional hardware. Software codes do not require hardware for compression or playback, but they typically do not deliver the same quality or compression ratio. Two of the major codec algorithms are:

• JPEG (Joint Photographic Experts Group). Originally developed for still images, JPEG can be compressed and decompressed at rates acceptable for nearly full-motion video (30 fps). JPEG still uses a series of still images, which is easier for editing. JPEG is typically lossy, but it can also be lossless. It eliminates redundant data for each individual image (intraframe). Compression efficiency is approximately 30:1 (20:1-40:1).
  
  
• MPEG (Motion Pictures Expert Group). Because MPEG can compress up to 200:1 at high-quality levels, it results in better, faster videos that require less space. MPEG is an interframe compressor. Because MPEG stores only incremental changes, it is not used during editing phases.

If you will be capturing, compressing, and playing video, you will need Microsoft Video for Windows (VFW) or QuickTime. Codecs are provided along with VFW:

• Cinepak. Although Cinepak can take longer to compress, it can produce better quality and higher compression than Indeo. It is also referred to as Compact Video Coded (CVC).
  
  
• Indeo. Indeo can outperform Cinepak and is capable of real-time compression. An Intel Smart Video board is required for real-time compression.
  
  
• Microsoft Video 1. Developed by MediaVision (MotiVE) and renamed MS Video 1, this codec is a DCT-based post-processor. A file is compressed after capture.

To play or record video on your multimedia PC (MPC), you will need some extra hardware and software:

• Video system software, such as Apple's QuickTime for Windows or Microsoft's Video for Windows.
  
  
• A compression/digitization video card that enables you to digitize and play large video files.
  
  
• An TV-to-VGA adapter that combines TV signals with computer video signals for output to a VCR. Video can come from a variety of sources: TV, VCR, video camera, or laserdisc player. When an animation file is recorded, it can be saved in a variety of different file formats: AVI (Audio Video Interleave), FLI (a 320x200 pixel animation file), or FLC (an animation file of any size).

You can incorporate these files into a multimedia presentation by using authoring software such as Icon Author from AIMTECH, or you can include the animated files as OLE objects to be used with Microsoft Word, Excel, Access, or other OLE-compliant applications.

When you connect video devices, use the S-Video (S-VHS) connector whenever available. This cable provides the best signal because separate signals are used for color (chroma) and brightness (luma). Otherwise, you will have to use composite video, which mixes luma and chroma. This results in a lower-quality signal. The better your signal, the better your video quality will be.

3-D Graphics Accelerators

A three-dimensional--or 3-D--image can contain an immense amount of detail. To manage that detail, 3-D application programs usually store and work with abstractions of the images, rather than the actual images themselves. Earlier, 3-D applications had to rely on support from software routines to convert these abstractions into live images. This heavily burdens the CPU in your PC, which significantly impacts the performance not only of the visual display but also of any other applications which may be running. Later there came a new breed of video accelerator chipsets found on many video adapters which could take on the job of rendering 3-D images, greatly lessening the load on your CPU and therefore greatly improving overall system performance.

The basic function of 3-D software is to convert image abstractions into what is seen on the video display. These abstractions generally consist of:

• Vertices. Locations in 3-D space, described in terms of their X, Y, and Z coordinates.
  
  
• Primitives. Simple geometric objects, described in terms of the relative locations of their vertices.
  
  
• Textures. Two-dimensional images or surfaces designed to be mapped onto primitives.

These abstract image descriptions must then be rendered, which means to be converted to visible form. Rendering depends on fairly standardized functions to convert the abstractions into the image that is displayed on-screen. The standard functions performed in rendering are:

• Geometry. Sizing, orienting, and moving primitives in space and calculating the effects of light sources.
  
  
• Rasterization. Converting primitives into pixels on the video display by filling the primitives with properly illuminated shading, textures, or a combination of the two.

Most video cards which are based on a chipset capable of 3-D video acceleration will have special built-in hardware that can perform the rasterization much faster than if done by software alone. Most chipsets with 3-D acceleration perform the following rasterization functions:

• Scan conversion. Determining which on-screen pixels are covered by each primitive.
  
  
• Shading. Filling pixels with colors that flow smoothly between the vertices.
  
  
• Texturing. Filling pixels with images from a 2-D picture or surface image.
  
  
• Visible surface determination. Identifying which pixels in a scene are obscured by objects closest to the viewer.
  
  
• Animation. Switching rapidly and cleanly to successive frames of motion sequences.

Hardware-accelerated rendering gives better image quality and faster animation than with software alone. Using special drivers, these cards can take over the intensive calculations formerly done by software to render a 3-D graphics image. This is especially useful if you are working with 3-D images, or if you are into the many games that rely extensively on 3-D displays.

System Video Information

There's more to video cards and monitors than just resolutions. You also have to know how the video card communicates to the monitor, and vice versa. The following sections describe how information is communicated to the monitor and back.

Monitor ID Pins

Table 10.26 shows the settings used for the monitor ID bits for several different IBM displays. By sensing which of these four pins are grounded, the video adapter can determine what type of display is attached. This is used especially with monochrome or color display detection. In this manner, the VGA or XGA circuitry can properly select the color mapping and image size to suit the display.

Table 10.26  IBM Display Monitor ID Settings

APM is a specification created by Microsoft and Intel that allows the system BIOS to manage the power consumption of the system and various system devices.

For displays, power management is implemented by a standard called DPMS (Display Power Management Signalling). This standard defines a method for signalling the monitor to enter into the various APM modes. The basis of the DPMS standard is the condition of the synchronization signals being sent to the display. By altering these signals, a DPMS-compatible monitor can be forced into the various APM modes.

The defined monitor states in DPMS are as follows:

• On. Refers to the state of the display when it is in full operation.
  
  
• Standby. Defines an optional operating state of minimal power reduction with the shortest recovery time.
  
  
• Suspend. Refers to a level of power management in which substantial power reduction is achieved by the display. The display can have a longer recovery time from this state than from the Standby state.
  
  
• Off. Indicates that the display is consuming the lowest level of power and is non-operational. Recovery from this state may optionally require the user to manually power on the monitor.

Table 10.27 summarizes the DPMS modes.

Table 10.27  Display Power Management Signaling

Solving most graphics adapter and monitor problems is fairly simple, although costly, because replacing the adapter or display is the usual procedure. A defective or dysfunctional adapter or display usually is replaced as a single unit, rather than repaired. The documentation required to service the adapters or displays properly is not always available. You cannot get schematic diagrams, parts lists, wiring diagrams, and so on for most of the adapters or monitors. Most adapters are constructed with surface-mount technology that requires a substantial investment in a rework station before you can remove and replace these components by hand.

Servicing monitors is a slightly different proposition. Although a display often is replaced as a whole unit, many displays are simply too expensive to replace. Your best bet is to either contact the company from which you purchased the display, or to contact one of the companies that specializes in monitor depot repair.

Depot repair means that you would send in your display to depot repair specialists who would either fix your particular unit or return an identical unit they have already repaired.

Troubleshooting a failed monitor is relatively simple. If your display goes out, for example, a swap with another monitor can confirm that the display is the problem. If the problem disappears when you change the display, then the problem was almost certainly in the original display; if the problem remains, then it is likely in the video card or PC itself.

---

CAUTION: You should not attempt to repair a display yourself. Touching the wrong item can be fatal. The display circuits sometimes hold extremely high voltages for hours, days, or even weeks after the power is shut off. A qualified service person should discharge the cathode ray tube and power capacitors before proceeding.

---

For most displays, you are limited to making simple adjustments. For color displays, the adjustments can be quite formidable if you lack experience. Even factory service technicians often lack proper documentation and service information.

If you have a problem with a display or adapter, it pays to call the manufacturer who might know about the problem and make repairs available, as occurred with the IBM 8513 display. Large numbers of the IBM 8513 color displays were manufactured with components whose values change over time and may exhibit text or graphics out of focus. IBM replaced these displays at no cost when focusing was a problem.

Remember that most of the problems you have with newer video adapters and displays will be related to the drivers that control these devices rather than the hardware itself. Contact the manufacturers to ensure that you have the latest and proper drivers; there may be a solution that you are unaware of.

DisplayMate

DisplayMate is a unique diagnostic and testing program that is designed to thoroughly test your video adapter and display. It is somewhat unique in that most conventional PC hardware diagnostics programs do not emphasize video testing as this program does. It is useful not only in testing if a video adapter is functioning properly, but also in checking out video displays. You can easily test the image quality of a display, which allows you to make focus, centering, brightness and contrast, color level, and other adjustments much more accurately than before. If you are purchasing a new display, you can use Display Mate to evaluate the sharpness and linearity of the display, and to have a consistent way of checking each one you are considering. If you use projection systems for presentations, you will find it invaluable for setting up and adjusting the projector.

DisplayMate can also test a video adapter thoroughly. It will set the video circuits into each possible video mode, and you can test what modes your card is capable of. It will also help you determine the performance level of your card, both with respect to resolution and colors as well as to speed. The program can be used to benchmark the performance of the display, which enables you to compare one type of video adapter system to another.